A morpheme analysis unit sets beforehand a meaning-candidate tag and a sentimental theme tag for a morpheme required to be input as a text. A syntax analysis unit generates an index where a clause including a meaning-candidate tag and a sentimental theme tag and a type of each tag. A meaning attribute extraction unit recognizes a clause including a meaning-candidate and a type of tag with reference to the index, and then applies a meaning attribute rule, sets a meaning attribute tag for a necessary clause, and updates the index. A sentimental analysis unit also recognizes a clause including a sentimental theme tag and a clause including a meaning attribute tag with reference to the index, and then applies a sentimental analysis rule and sets a sentimental attribute tag for a necessary clause.
Legal claims defining the scope of protection, as filed with the USPTO.
1. A text analysis system, comprising: a memory configured to store attribute dictionaries; and a processor configured to decompose input text into morphemes; set pre-tags in the memory for the morphemes with reference to the attribute dictionaries, each pre-tag specifying a correspondence relationship between specific morphemes and an attribute type; identify a dependency relationship between respective morphemes or respective clauses, each clause being an aggregation of at least two of the morphemes; generate an index, in the memory, combining identification information that identifies a first one of the morphemes or the clauses, the pre-tag set for the first one of the morphemes or the clauses, and a tag type of the pre-tag; store a plurality of determination rules, in the memory, each including an application condition that designates at least a second one of the morphemes or the clauses, including a specific type of the pre-tags, and an application effect that specifies a third one of the morphemes or the clauses for which an attribute tag is set and the attribute type of the attribute tag; and set, with reference to the index and the determination rules, the attribute tag of the attribute type of at least one concerned morpheme or at least one concerned clause in a concerned text, designated in one of the determination rules when the application effect and the application condition of the one of the determination rules match the at least one concerned morpheme or the at least one concerned clause in the concerned text.
2. The text analysis system according to claim 1 , wherein: at least one of the attribute dictionaries is a meaning attribute dictionary that specifies a first correspondence relationship between first specific morphemes and the attribute type of a meaning attribute; a meaning-candidate tag is set as the pre-tag for a first concerned morpheme; the index includes a first combination of the identification information that identifies a first clause having the meaning-candidate tag and the tag type of the meaning-candidate tag; the determination rules include a second combination of a first application condition that designates at least first clauses having a first specific type of the meaning-candidate tag and a first application effect that designates second clauses having the meaning attribute tag and the attribute type of the meaning attribute tag; and the system determines, with reference to the index and the determination rules, the meaning attribute tag of the attribute type designated in the second combination when the first application effect and the first application condition matches the at least one concerned clause in the concerned text.
3. The text analysis system according to claim 2 , wherein: at least another of the attribute dictionaries is a sentimental theme dictionary that specifies a second correspondence relationship between the specific morphemes and a theme type of a sentimental theme; a sentimental theme tag is set as the pre-tag for a second concerned morpheme; the index includes a third combination of the identification information that identifies a second clause having the sentimental theme tag and the theme type of the sentimental theme tag; the determination rules include a fourth combination of a second application condition that designates third clauses having a second specific type of the sentimental theme tag and the attribute type of the meaning attribute tag included in the second clause or a third clause, and a second application effect that designates fourth clauses having a sentimental attribute tag and the attribute type of the sentimental attribute tag; and the system determines, with reference to the index and the determination rules, the sentimental attribute tag of the attribute type designated in the fourth combination when the second application effect and the second application condition matches the at least one concerned clause in the concerned text.
4. The text analysis system according to claim 1 , wherein: the attribute dictionaries include a sentimental theme dictionary that specifies a first correspondence relationship between the specific morphemes and a theme type of a sentimental theme, and a large/small expression dictionary that specifies a second correspondence relationship between the specific morphemes and an expression type of a large/small expression; a sentimental theme tag and a large/small expression tag are set as pre-tags for a concerned morpheme; the index includes a first combination of the identification information that identifies a first clause having the sentimental theme tag and the theme type of the sentimental theme tag, and a second combination of the identification information that identifies a second clause having the large/small expression tag and the expression type of the large/small expression; the determination rules include a third combination of a first application condition that designates third clauses having a first specific type of the sentimental theme tag and a second specific type of the large/small expression tag included in the third clauses or a different clause, and a first application effect that designates fourth clauses having the sentimental attribute tag and the attribute type of the sentimental attribute tag; and the system determines, with reference to the index and the determination rules, the sentimental attribute tag of the attribute type designated in the first application effect when the first application condition matches the at least one concerned morpheme or the at least one concerned clause in the concerned text.
5. The text analysis system according to claim 1 , wherein: the attribute dictionaries include a sentimental theme dictionary that specifies a first correspondence relationship between the specific morphemes and a theme type of a sentimental theme, and a sentimental expression dictionary that specifies a second correspondence relationship between the specific morphemes and an expression type of a sentimental expression; a sentimental theme tag and a sentimental expression tag are set as pre-tags for a concerned morpheme; the index includes a first combination of the identification information that identifies a first clause having the sentimental theme tag and the theme type of the sentimental theme tag, and a second combination of the identification information that identifies a second clause having the sentimental expression tag and the expression type of the sentimental expression; the determination rules include a third combination of a specific application condition that designates third clauses having a first specific type of the sentimental theme tag and a second specific type of the sentimental expression tag included in the third clauses or a different clause, and a specific application effect that designates fourth clauses having the sentimental attribute tag and the attribute type of the sentimental attribute tag; and the system determines, with reference to the index and the determination rules, the sentimental attribute tag of the attribute type designated in the specific application effect when the specific application condition matches the at least one concerned morpheme or the at least one concerned clause in the concerned text.
6. A non-transitory computer readable medium embodying a text analysis program which, when executed by a computer, causes the computer to perform a method comprising: decomposing input text into morphemes; setting pre-tags for the morphemes with reference to attribute dictionaries each pre-tag specifying a correspondence relationship between specific morphemes and an attribute type; identifying a dependency relationship between respective morphemes or respective clauses, each clause being an aggregation of at least two of the morphemes; generating an index containing a first combination of identification information that identifies a first one of the morphemes or the clauses, the pre-tag set for the first one of the morphemes or the clauses, and a pre-tag type of the pre-tag; storing a plurality of determination rules, each including a second combination of an application condition that designates at least a second one of the morphemes or the clauses, including a specific type of the pre-tags, and an application effect that specifies a third one of the morphemes or the clauses for which an attribute tag is set and the attribute type of the attribute tag; and determining, with reference to the index and the determination rules, the attribute tag of the attribute type of at least one concerned morpheme or at least one concerned clause in a concerned text, designated in one of the determination rules when the application effect and the application condition of the one of the determination rules match the concerned text.
7. The text analysis system according to claim 4 , wherein: the attribute dictionaries further include a sentimental expression dictionary that specifies a third correspondence relationship between the specific morphemes and an expression type of a sentimental expression; a sentimental expression tag is set as another pre-tag for the concerned morpheme; the index includes a fourth combination of the identification information that identifies a fifth clause having the sentimental theme tag and the theme type of the sentimental theme tag, and a fifth combination of the identification information that identifies a sixth clause having the sentimental expression tag and the expression type of the sentimental expression; the determination rules include a sixth combination of a second application condition that designates seventh clauses having a third specific type of the sentimental theme tag and a fourth specific type of the sentimental expression tag included in the seventh clauses or a different clause, and a second application effect that designates eighth clauses having the sentimental attribute tag and the attribute type of the sentimental attribute tag; and the system determines, with reference to the index and the determination rules, the sentimental attribute tag of the attribute type designated in the second application effect when the second application condition matches the at least one concerned morpheme or the at least one concerned clause in the concerned text.
Cooperative Patent Classification codes for this invention. Click any code to explore related patents in that topic.
September 21, 2018
November 17, 2020
Browse 5M+ US patents with plain-English claim translations and AI-generated analysis.